Measurement-based Channel Characterization for A2A and A2G Wireless Drone Communication Systems

This paper presents field measurement-based channel characterization for air--to--ground (A2G) and air--to--air (A2A) wireless communication systems using two drones equipped with lightweight software-defined radios. A correlation-based channel sounder is employed such that the transmitting drone broadcasts the sounding waveform with a pseudo-noise sequence and the receiving drone captures the sounding waveform together with the location information for the post-processing analysis. The path loss results demonstrate that the measurement and flat-earth two-ray results have similar trends for A2G while the measurement and free space path loss are similar to each other for A2A. The time delays between the direct path and multipath components are widely spread for A2A while the multipath components are mostly concentrated around the direct path for A2G generating a more challenging communication environment. We observe that the reflections from several buildings having metal roofs and claddings on the measurement site cause sudden peaks in the root-mean-square delay spread. The results indicate that the A2A channel has better characteristics than the A2G under similar mobility conditions.

Deep Convolutional Learning-Aided Detector for Generalized Frequency Division Multiplexing with Index Modulation

In this paper, a deep convolutional neural network-based symbol detection and demodulation is proposed for generalized frequency division multiplexing with index modulation (GFDM-IM) scheme in order to improve the error performance of the system. The proposed method first pre-processes the received signal by using a zero-forcing (ZF) detector and then uses a neural network consisting of a convolutional neural network (CNN) followed by a fully-connected neural network (FCNN). The FCNN part uses only two fully-connected layers, which can be adapted to yield a trade-off between complexity and bit error rate (BER) performance. This two-stage approach prevents the getting stuck of neural network in a saddle point and enables IM blocks processing independently. It has been demonstrated that the proposed deep convolutional neural network-based detection and demodulation scheme provides better BER performance compared to ZF detector with a reasonable complexity increase. We conclude that non-orthogonal waveforms combined with IM schemes with the help of deep learning is a promising physical layer (PHY) scheme for future wireless networks